POS0133 THE FINDINGS FROM MOST RANDOMIZED CONTROLLED TRIALS OF GIANT CELL ARTERITIS AND TAKAYASU ARTERITIS ARE FRAGILE

Background: The fragility of randomized controlled trials (RCTs) can be described using the fragility index, which is the minimum number of events that need to be have a different outcome in the intervention group to render a trial with significant results as non-significant (fragility index - FI) and vice versa (reverse fragility index – RFI) using Fisher's exact test [1,2]. No studies to date have described the FI of trials in large vessel vasculitis. Objectives: To evaluate the robustness of published RCTs in giant cell arteritis (GCA) and Takayasu arteritis (TAK) identified via a systematic review by assessing their fragility (FI or RFI) and fragility quotient (FQ) for trials with categorical outcomes. Methods: RCTs that evaluated GCA or TAK were identified using a systematic review of MEDLINE (via Ovid), EMBASE, CENTRAL, clinicaltrials.gov and ISRCTN on 9 th December 2024 (PROSPERO registration identifier CRD42025619443). The FI or RFI were calculated for primary or key secondary endpoints (if primary endpoint was not categorical). FQ was calculated as the ratio of FI (or RFI) to total number of trial subjects, reflecting what percentage of trial participants needed to have a different outcome for a significant trial to become non-significant (or vice versa). Some trials had more than one primary endpoint. Risk of Bias (High, Some Concern, Low) was assessed using Cochrane Risk of Bias 2 tool. The FQ was compared between studies with low risk of bias versus those with some concerns or high risk of bias using unpaired Student's t-test. Results: From 3874 records screened after deduplication, 30 primary reports of RCTs were identified (GCA:n=23; TAK:n=7), from which 16 trials with categorical outcomes (GCA: n=12; TAK: n=4) published after 2000 were analyzed for FI or RFI and FQ. For trials with a significant primary end-point in GCA [n=6, RoB low for 3], the FI ranged from 1 to 12 and FQ ranged from 0.014 to 0.115. The one significant trial in TAK had FI of 1 and FQ 0.025 (Table 1). For trials without significant primary end-points in GCA [n=7, RoB low for 2], the RFI ranged from 1 to 9 and FQ ranged from 0.015 to 0.33. For trials without a significant primary end-point in TAK [n=3, RoB low for 1], RFI ranged from 2 to 6 and FQ ranged from 0.076 to 0.140 (Table 2). FQ was similar for trials with low risk of bias or those with some concerns or high risk of bias (mean±SD 0.10±0.04 vs 0.09±0.09, p=0.884). FQ was ≤0.1 for 9/12 trials of GCA and 2/4 trials of TAK. Conclusion: Most published trials in large vessel vasculitis are fragile. Including larger numbers of patients in future RCTs of GCA and TAK would increase the robustness of their results. REFERENCES: [1] Figueroa-Parra G, et al Lupus Sci Med. 2024;11(1):e001068. [2] Khan MS, et al. JAMA Netw Open. 2020 Aug 3;3(8):e2012469. Table 1Assessment of fragility index of randomized controlled trials of GCA and TAK published after 2000 with significant primary endpoints.DiseaseAuthor YearPrimary endpointEvents in intervention vs comparator group: Fragility index (FI)FQRoBGCAHoffman 2002First relapse by 12 monthsMethotrexate 31/39 vs PBO 31/32:FI=10.014HMazlumzadeh 2006In remission on ≤5 mg prednisolone at 36 weeksIVMP 10/14 vs PBO 2/13:FI=20.074HVilliger 2016Complete remission at prednisolone 0.1 mg/kg/day at week 12Tocilizumab 17/20 vs PBO 4/10:FI=10.033SStone 2017Rate of sustained glucocorticoid-free remission at week 52Tocilizumab weekly 56/100 vs PBO 7/50:FI=12Tocilizumab two-weekly 26/49 vs PBO 7/50:FI=100.080.10LCid 2022Sustained remission at week 26 (key secondary end-point)Mavrilimumab 35/42 vs PBO 14/28:FI=30.043LVenhoff 2023Sustained remission at week 28Secukinumab 16/27 vs PBO 2/25:FI=60.115LTAKWang 2024Efficacy rate at 6 months; Complete remission at 6 monthsAdalimumab 18/21 vs Tocilizumab 10/19 for both:FI=10.025HIVMP – Intravenous methylprednisolone pulse; PBO – Placebo.RoB – Risk of Bias; L – Low, S – Some concerns, H-High. Table 2Assessment of reverse fragility index of randomized controlled trials of GCA and TAK published after 2000 with non-significant primary end-pointsDiseaseAuthor YearPrimary endpointEvents in intervention vs comparator group: Reverse Fragility Index (RFI)FQRoBGCAChevalet 2000On glucocorticoids at 1 year*IVMP + 0.7 mg/kg/d prednisone 46/54 vs 0.7 mg/kg/d prednisone 33/43:RFI=4IVMP + 0.7 mg/kg/d prednisone 46/54 vs IVMP + 0.5 mg/kg/d prednisone 32/36:RFI=80.7 mg/kg/d prednisone 33/43 vs IVMP + 0.5 mg/kg/d prednisone 32/36:RFI=40.0240.0490.024HJover 2001Disease relapsesMethotrexate 7/15 vs PBO 15/18:RFI=10.024HHoffman 2002**First relapse by 6 monthsTreatment failure by 6 monthsTreatment failure by 12 monthsMethotrexate 29/42 vs PBO 24/35:RFI=9Methotrexate 10/39 vs PBO 12/33:RFI=5Methotrexate 22/37 vs PBO 24/29:RFI=10.1170.0690.015HHoffman 2007Relapse-free at 22 weeksAdverse eventsInfliximab 12/28 vs PBO 8/16:RFI=7Infliximab 26/28 vs PBO 15/16:RFI=80.1590.182LMartinez-Taboada 2007Withdrawal of corticosteroids with control of disease activity at 12 monthsEtanercept 4/8 vs PBO 2/9:RFI=20.118HSeror 2013Percentage in remission on <0.1 mg/kg prednisone at 26 weeksAdalimumab 20/34 vs PBO 18/36:RFI=70.1LRaine 2018Persistent clinical disease control at 26 weeksMR prednisone 6/7 vs IR prednisone 4/5:RFI=40.33STAKLangford 2017Relapse-free survival at month 12 (ITT)Abatacept 2/11 vs PBO 6/15:RFI=20.076LNakaoka 2018Relapses (ITT)Tocilizumab 8/18 vs PBO 11/18:RFI=40.111SPadiyar 2024Clinical response at 9 monthsMycophenolate 15/21 vs Methotrexate 14/22:RFI=60.140S* One out of five primary end-points was categorical.**Had more than one primary endpoint.IR – Immediate release, IVMP – Intravenous methylprednisolone pulse, MR – Modified release; PBO – Placebo.RoB – Risk of Bias: L – Low, S – Some concerns, H-High. Acknowledgements: None . Disclosure of Interests: None declared . © The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Neither EULAR nor the publisher make any representation as to the accuracy of the content. The authors are solely responsible for the content in their abstract including accuracy of the facts, statements, results, conclusion, citing resources etc.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.